Evidence for wing development in the Late Palaeozoic Palaeodictyoptera revisited

Thoracic and abdominal outgrowths in early pterygotes: a clue to the common ancestor of winged insects?

One of the fundamental questions in insect evolution is the origin of their wings and primary function of ancestral wing precursors. Recent phylogenomic and comparative morphological studies broadly support a terrestrial ancestor of pterygotes, but an aquatic or semiaquatic ancestor cannot be ruled out. Here new features of the branchial system of palaeodictyopteran larvae of several different instars of Katosaxoniapteron brauneri gen. et sp. nov. (Eugereonoidea) from the late Carboniferous collected at Piesberg (Germany) are described, which consist of delicate dorsolateral and lamellate caudal abdominal gills that support an aquatic or at least semiaquatic lifestyle for these insects. Moreover, the similar form and surface microstructures on the lateral abdominal outgrowths and thoracic wing pads indicate that paired serial outgrowths on segments of both tagmata presumably functioned as ancestral type of gills resembling a protopterygote model. This is consistent with the hypothesis that the wing sheaths of later stage damselfly larvae in hypoxic conditions have a respiratory role similar to abdominal tracheal gills. Hence, the primary function and driving force for the evolution of the precursors of wing pads and their abdominal homologues could be respiration.

The function of wing bullae in mayflies (Insecta: Ephemeroptera) reveals new insights into the early evolution of Pterygota

BACKGROUND

Mayflies are basal winged insects of crucial importance for the understanding of the early evolution of Pterygota. Unlike all other insects, they have two successive winged stages, the subimago and the imago. Their forewings feature so-called bullae, which are desclerotized spots in the anterior main veins. Up to now, they have been considered to play a major role in wing bending during flight.

RESULTS

We investigated bullae by multiple methods to reveal their structure and arrangement and to gain new information on the evolution of insect flight. Bullae are mostly present in the anterior negative wing veins, disrupting the otherwise rigid veins. High-speed videography reveals that mayfly wings do not bend during flight. Likewise, different arrangements of bullae in different species do not correlate with different modes of flying. Observations on the moulting of subimagines unravel that they are essential for wing bending during the extraction of the imaginal wing from the subimaginal cuticle. Bullae define predetermined bending lines, which, together with a highly flexible wing membrane enriched with resilin, permit wing bending during subimaginal moulting. Bullae are only absent in those species that remain in the subimaginal stage or that use modified modes of moulting. Bullae are also visible in fossil mayflies and can be traced back to stemgroup mayflies of the Early Permian, the 270 million years old Protereismatidae, which most probably had bullae in both fore- and hind wings.

CONCLUSIONS

Bullae in mayfly wings do not play a role in flight as previously thought, but are crucial for wing bending during subimaginal moulting. Thus, the presence of bullae is a reliable morphological marker for a subimaginal life stage, confirming the existence of the subimago already in Permian Protereismatidae. A thorough search for bullae in fossils of other pterygote lineages may reveal wheather they also had subimagines and at what point in evolution this life stage was lost. In mayflies, however, the subimago may have been retained due to selective advantages in connection with the transition from aquatic to terrestrial life or due to morphological requirements for a specialized mating flight.

Diversity, Form, and Postembryonic Development of Paleozoic Insects

While Mesozoic, Paleogene, and Neogene insect faunas greatly resemble the modern one, the Paleozoic fauna provides unique insights into key innovations in insect evolution, such as the origin of wings and modifications of postembryonic development including holometaboly. Deep-divergence estimates suggest that the majority of contemporary insect orders originated in the Late Paleozoic, but these estimates reflect divergences between stem groups of each lineage rather than the later appearance of the crown groups. The fossil record shows the initial radiations of the extant hyperdiverse clades during the Early Permian, as well as the specialized fauna present before the End Permian mass extinction. This review summarizes the recent discoveries related to the documented diversity of Paleozoic hexapods, as well as current knowledge about what has actually been verified from fossil evidence as it relates to postembryonic development and the morphology of different body parts.

Abdominal serial homologues of wings in Paleozoic insects

The Late Paleozoic acquisition of wings in insects represents one of the key steps in arthropod evolution. While the origin of wings has been a contentious matter for nearly two centuries, recent evolutionary developmental studies suggest either the participation of both tergal and pleural tissues in the formation of wings¹ or wings originated from exites of the most proximal leg podite incorporated into the insect body wall.² The so-called "dual hypothesis" for wing origins finds support from studies of embryology, evo-devo, and genomics, although the degree of the presumed contribution from tergal and pleural tissues differ.^3-6 Ohde et al.,⁷ confirmed a major role for tergal tissue in the formation of the cricket wing and suggested that "wings evolved from the pre-existing lateral terga of a wingless insect ancestor." Additional work has focused on identifying partial serially homologous structures of wings on the prothorax^8,9 and abdominal segments.¹⁰ Thus, several studies have suggested that the prothoracic horns in scarab beetles,⁹ gin traps of tenebrionid and scarab beetle pupae,^11,12 or abdominal tracheal gills of mayfly larvae^1,13 evolved from serial homologues of wings. Here, we present critical information from abdominal lateral outgrowths (flaps) of Paleozoic palaeodictyopteran larvae, which show comparable structure to thoracic wings, consisting of cordate lateral outgrowths antero-basally hinged by muscle attachments. These flaps therefore most likely represent wing serial homologues. The presence of these paired outgrowths on abdominal segments I-IX in early diverging Pterygota likely corresponds to crustacean epipods^14,15 and resembles a hypothesized ancestral body plan of a "protopterygote" model.